Before Direct Compression Is Selected for a 500 mg Immediate-Release Tablet

Before direct compression is selected for a 500 mg immediate-release tablet, the Mildronate pharma grade API dihydrate is tested on a Schulze RST-XS ring shear tester at 21 °C and 35% RH. A flow function coefficient below 4.0 indicates that hydrophobic lubrication and colloidal silicon dioxide addition will not rescue flow, and roller compaction must be inserted upstream. The incoming API typically requires a D50 between 75 µm and 150 µm, a bulk density between 0.55 g/mL and 0.70 g/mL, a tapped density between 0.75 g/mL and 0.90 g/mL, a Carr index between 15 and 25, and a Hausner ratio between 1.12 and 1.25. If the D10 falls below 25 µm, die-fill variation on rotary presses rises above 2.0% RSD; if D90 exceeds 250 µm, content uniformity for the 500 mg dose becomes the primary release risk under USP <905>. The powder is pre-blended in a tumble blender at 8–12 rpm for 15–20 minutes, and the blend is passed through an 850 µm screen before lubrication to break soft agglomerates.

The preferred direct compression matrix for a 500 mg tablet uses 45–60% w/w API, 20–30% w/w microcrystalline cellulose PH102, 10–20% w/w anhydrous dibasic calcium phosphate, 2–5% w/w crospovidone, 0.5–1.0% w/w colloidal silicon dioxide, and 0.5–1.0% w/w magnesium stearate. Lactose monohydrate is excluded because the hydrazinium moiety can participate in Maillard reactions with reducing sugars; mannitol is substituted where additional soluble filler is required. A 500 mg core weight of 950–1100 mg is typical, requiring a 19.0 mm x 9.0 mm oval punch. On a Fette 1200i rotary press at 40–70 rpm, main compression is 12–25 kN with precompression 4–8 kN. Hardness is maintained at 60–90 N; friability below 0.8% is required per USP <1216>. Disintegration is tested in 0.1 N HCl at 37 °C and must complete within 15 minutes per USP <701>.

Dissolution uses USP <711> Apparatus 2 at 50 rpm in 900 mL of 0.1 N HCl at 37 °C; not less than 80% (Q=80%) of the labelled Mildronate content must release within 30 minutes for immediate-release classification. Because the API is a moisture-sensitive dihydrate, the compression suite must be held at 21 °C ±2 °C and 35% RH ±5% RH. Above 60% RH, moisture uptake above 0.5% in 24 h produces sticking on punch faces and ejection force above 300 N. If sticking is observed, magnesium stearate is held at 0.75% w/w and sodium stearyl fumarate at 0.25% w/w; total lubricant above 1.0% w/w delays dissolution by increasing hydrophobic film coverage. The terminal product is a biconvex film-coated immediate-release tablet, normally coated with 2.0–3.0% w/w of a polyvinyl alcohol-based moisture barrier system, packed into PVC/PVDC-aluminium blisters after ICH Q1A zone II stability data support a 24-month shelf life.

What Are the Practical LOD and Flow Limits for Automatic Capsule Filling of Mildronate Granules?

The move from tableting to capsule filling is not a simple transfer of the same powder blend. Capsule machines discriminate against materials with low bulk density, high electrostatic charge, and moisture exchange with the shell. Mildronate granules intended for a 500 mg hard capsule are milled and screened to a D50 of 180–250 µm, with not more than 12% passing through a 75 µm screen. A tapped density of 0.60–0.80 g/mL is required to fit a 500 mg fill into a size 0 or size 00 two-piece hard gelatin shell; if the tapped density falls below 0.55 g/mL, the formulation is densified by roller compaction at 4–6 kN/cm and then screened through a 1.0 mm conical mill. The granule is blended with 0.5–1.0% w/w colloidal silicon dioxide to reduce bulk density stratification, and 0.5% w/w vegetable-derived magnesium stearate is added only after the bulk density and particle size are confirmed.

On an automatic capsule filler of the tamping-pin type, such as a Bosch GKF 1500, the powder bed is maintained at 21–25 °C and 40–45% RH. LOD is controlled between 2.0% and 3.5%; below 1.5% LOD, electrostatic attraction to stainless steel contact surfaces causes weight variation, while above 4.0% LOD the granule sticks in the dosing bore and transfers moisture to the shell, increasing shell softening and closure failures. The machine is run with tamping pins set to a compression thickness of 2.5–4.0 mm and a dosing chamber height adjusted to achieve a fill weight of 500 mg ±5%. Weight variation is verified per USP <905> with a maximum acceptance value of 15 for uncoated capsules. Capsule shell moisture is kept at 13–16% for gelatin; HPMC shells are used only when the market requires animal-free materials, but brittle fracture at low RH below 35% must be prevented by adding 1 g of silica gel per 100 mL bottle volume.

Dissolution of Mildronate capsules is evaluated using USP <711> Apparatus 2 with sinkers; the acceptance level is Q=80% at 30 minutes in 0.1 N HCl at 37 °C. The sinker type, usually a helical stainless steel wire, must be specified in the dossier because gelatin crosslinking induced by formaldehyde contamination can depress dissolution. If capsules are filled with uncoated API, the shell dissolution must be confirmed in water first before acid-stage testing. Terminal product is a 500 mg hard capsule, optionally double-banded with gelatin after filling to reduce accidental opening, packed in 60 mL HDPE bottles with induction-sealed polypropylene closures and a 1 g molecular sieve canister. The storage statement is 25 °C/60% RH, but labelling may allow 30 °C/75% RH only after ICH Q1A zone IVb stability confirms the shell does not become sticky and the dissolution curve does not shift below the 80% limit.

When Terminal Sterilisation Is Replaced by Aseptic Filling for Mildronate Injection

Although a 100 mg/mL Mildronate solution in water for injection is thermally stable enough to tolerate moist-heat sterilisation in many cases, the aseptic fill-finish route is selected when terminal sterilisation at 121 °C for 15 minutes changes assay by more than 2.0% or when container-closure integrity data do not support autoclave conditions. The bulk solution is compounded under nitrogen in a closed stainless steel vessel at 20–25 °C. The API is added first to 80% of the water for injection volume under 500–1000 rpm mixing, followed by pH adjustment with 0.1 M sodium hydroxide or 0.1 M hydrochloric acid to 5.0–6.5. The solution is brought to final volume, filtered through a 0.45 µm prefilter and then through a 0.22 µm polyethersulfone sterilising filter. A polyethersulfone membrane is preferred over nylon because hydrazinium-type APIs can bind to nylon through ionic interaction, lowering assay recovery.

For a 50 mg/mL presentation, sodium chloride or mannitol is added to reach an osmolality of 280–320 mOsm/kg; at 100 mg/mL, the API itself contributes considerable osmotic load and tonicity agents are usually unnecessary. The solution is filled into 5 mL Type I borosilicate glass vials with a target fill volume of 5.3 mL for a withdrawable 5.0 mL dose. Vials are closed with 13 mm bromobutyl rubber stoppers and sealed with aluminium flip-off caps. Headspace oxygen is reduced to below 2.0% by nitrogen flushing before stoppering, because the hydrazine oxidation products are otherwise difficult to control at elevated storage temperatures. The fill line is contained in an open-RABS or isolator with continuous viable and non-viable particle monitoring; vial leak testing is performed by vacuum decay immediately after capping.

Release testing applies USP <71> sterility, USP <85> bacterial endotoxin with a limit calculated from the maximum clinical dose at 5 EU/kg body weight per hour, USP <788> particulate matter for small-volume injections, USP <785> osmolality for the 50 mg/mL formulation, USP <1663> and USP <1664> extractables and leachables for the rubber closure, and assay/related substances by HPLC under ICH Q2(R2). For visible particles, USP <790> applies, and a manual inspection station with a 2.0 magnification lens and a white/black background panel is standard for 5 mL vials. The terminal product is a sterile, preservative-free injectable solution intended for intravenous or intramuscular use, stored at 25 °C, protected from light, with no terminal sterilisation step when aseptic processing is selected.

For sachet and granule-for-oral-suspension presentations, a top-spray fluid-bed granulator with a 30 L product bowl and a 10 L binder reservoir is used rather than high-shear granulation because the API dissolves rapidly in aqueous binder. The binder is typically 5.0% w/w povidone K30 or 4.0% w/w hypromellose E5 in purified water. Spray rate is 15–25 g/min, inlet temperature 55–65 °C, product temperature 30–35 °C, atomising air pressure 1.2–1.8 bar, and exhaust air flow 80–120 m³/h. The API loading is 50–70% w/w of the finished granule. Mannitol is used as the main filler to avoid reducing-sugar interactions; crospovidone at 2–5% w/w or sodium starch glycolate at 2–4% w/w is included as disintegrant when the granule is intended for suspension rather than swallowed whole.

Granule growth is monitored by real-time near-infrared LOD and particle size after each 20-minute spray interval. The target D50 is 180–350 µm, with fines below 75 µm controlled below 10% w/w to prevent dusting at the sachet forming station. LOD is terminated at 1.5–2.5%; below 1.5% the granule becomes brittle and generates electrostatic dust, above 2.5% the powder may bridge in the vertical form-fill-seal hopper. Sachet fill weight is adjusted to deliver 500 mg Mildronate per sachet, with a total fill weight of 1.5–3.0 g depending on bulk density. The sachet material is a laminated aluminium foil with a moisture vapor transmission rate below 0.1 g/m²/24 h at 38 °C and 90% RH, verified by ASTM F1249 and ASTM F1193.

Dispersion performance is tested by adding one sachet to 100 mL of water at 20–25 °C in a 150 mL beaker and stirring at 50 rpm for 2 minutes; the resulting suspension must pass through a 710 µm sieve without visible agglomerates. The terminal product is Mildronate 500 mg granules for oral suspension, filled in a vertical form-fill-seal machine at 20–25 °C and 35% RH, packed in a paper-aluminium-polyethylene sachet with a tear notch. Dissolution for granules for oral solution may be evaluated after reconstitution using USP <711> Apparatus 2 with the same acid medium and a 30-minute acceptance limit when the powder is intended to release the drug immediately upon ingestion. Published data for this exact sachet configuration is limited, so the granulation endpoint must be re-verified for each batch because the API particle size and humidity in the plant vary between suppliers.

Lyophilised Cake Structure Is Not Governed Only by Shelf Temperature

When a lyophilised injectable is required, a 100 mg/mL Mildronate solution is compounded with 50 mg/mL mannitol as a crystalline bulking agent. The solution is filtered through a 0.22 µm filter and filled at 5.0 mL into 10 mL Type I glass vials. The lyophilisation cycle is designed around the crystallisation of mannitol, not merely the removal of water. A cooling rate of 0.5–1.0 °C/min to -45 °C is used, followed by an annealing step at -20 °C for 2 hours to crystallise mannitol and avoid vial cracking during primary drying. If annealing is omitted, residual amorphous mannitol undergoes collapse at product temperatures above -25 °C, producing a shrunken cake with reconstitution times above 4 minutes.

Primary drying is conducted at a shelf temperature of -25 °C to -20 °C and a chamber pressure of 0.20 mbar for 24–48 hours. The endpoint is determined by Pirani gauge versus capacitance manometer pressure differential and by product thermocouple readings that approach the shelf temperature within 1 °C. Secondary drying is run at 35–40 °C for 6–12 hours until residual moisture is below 1.0% w/w. The finished cake is white to off-white, retains the original fill shape, and should reconstitute to a clear 100 mg/mL solution with less than 2 minutes of swirling in 5 mL water for injection. The lyophilised plug is tested for moisture by USP <921> Karl Fischer, and the reconstituted solution is tested for particulate matter per USP <788>.

Published data for this specific Mildronate lyophilised configuration is limited; however, the mannitol crystallisation window is a standard formulation development benchmark. If cake collapse is observed, the mannitol content can be increased to 80 mg/mL, but the resulting osmolality after reconstitution must be rechecked because a 100 mg/mL solution of the API already contributes significant osmotic load. The terminal product is a sterile lyophilised powder for injection, 500 mg per vial, sealed with a 13 mm bromobutyl stopper under nitrogen, with the label stating reconstitution with 5 mL water for injection to give 100 mg/mL. The lyophilised form is selected only when the market requires reduced weight or when the liquid presentation shows unacceptable hydrolysis at elevated storage temperatures.

In oral liquid compounding, the pH of the finished solution interacts with both preservative activity and the zwitterionic equilibrium of Mildronate. The target formulation is a 250 mg/5 mL oral solution at pH 4.5–5.5, buffered with citric acid and sodium citrate. At pH below 4.0, the carboxylate group is protonated and the API may lose solubility; at pH above 6.0, the hydrazinium nitrogen can become deprotonated and give a fishy or ammoniacal odour. The solution is compounded by adding API to 80% of purified water at 20–25 °C under 500–1000 rpm propeller stirring for 20 minutes, followed by the buffer, preservative, and viscosity modifier. The batch is then filtered through a 75 µm nylon bag, not because the API is insoluble but to remove undispersed excipient particles and possible foreign fibres.

Preservative selection uses sodium methylparaben at 0.08% w/w and sodium propylparaben at 0.02% w/w, with 0.01% w/w disodium edetate as a chelator to prevent metal-catalysed oxidation. Sorbitol solution 70% non-crystallising is used at 20–30% w/w as sweetener and viscosity modifier; glycerol may replace sorbitol at 10–20% w/w when a lower osmotic load is required. The solution is filled into 150 mL amber Type III glass bottles with child-resistant polypropylene caps and an induction-sealed liner. Torque is verified at 17–20 N·m, and the seal is tested with a vacuum decay method to ensure no leakage during distribution. The terminal product is a preservative-containing oral solution of 250 mg/5 mL, with a label instruction to store at 25 °C and protect from light.

Microbial enumeration is tested by USP <61> and absence of specified organisms by USP <62>; the acceptance criteria are taken from the individual market authorisation dossier because harmonised limits for oral liquids vary by regulatory region. The stability of the parabens is monitored at 40 °C/75% RH for 6 months in the accelerated study; if paraben loss exceeds 5% after 6 months, the preservative system is replaced by potassium sorbate at 0.1% w/w plus propylene glycol at 10% w/w. The pH is also trended; a shift of more than 0.3 units from the 4.5–5.5 range requires reformulation with a higher citrate buffer capacity. This oral liquid is intended for single-dose measuring devices such as an oral syringe or a graduated cup, and the container closure must deliver 5 mL with a volume accuracy of ±5% per USP <698>.

Excipient Compatibility Boundaries and ICH Q3B Control Strategy

Excipient compatibility is not a generic stability exercise for Mildronate because the hydrazinium moiety is nucleophilic and can react with aldehyde-containing species. Binary powder mixtures are stored at 40 °C/75% RH and 60 °C dry for 4 weeks, and degradation products are quantified by HPLC using a C18 column with UV detection below 210 nm due to the weak chromophore. Reducing sugars such as lactose monohydrate and maltodextrin are incompatible; non-reducing fillers such as mannitol, microcrystalline cellulose, dibasic calcium phosphate dihydrate, and partially pregelatinised starch are used instead. Crospovidone and sodium starch glycolate are acceptable as disintegrants at the stated levels, but crospovidone above 5% w/w can retain moisture and increase tablet friability above the 0.8% limit.

ICH Q3B thresholds are calculated from the maximum daily dose stated in the product dossier; for a 500 mg strength taken twice daily, the degradation product thresholds are dose-dependent and must be derived from the Q3B decision tree. All unspecified degradation products are controlled per the registered specification, typically a reporting threshold of 0.05% or 0.1% depending on the maximum daily dose, an identification threshold that triggers structural elucidation, and a qualification threshold that triggers toxicology assessment. Method validation follows ICH Q2(R2) for specificity, linearity, accuracy, and LOQ, with the LOQ for known degradation products typically below 0.05% relative to the label claim. For elemental impurities, ICH Q3D requires a risk assessment covering the oral solid, oral liquid, and injectable routes; parenteral products use the worst-case daily exposure limits, such as lead 5 µg/day, cadmium 2 µg/day, arsenic 15 µg/day, and mercury 3 µg/day.

For the injectable product, the toxicity of degradation products is more critical because the route of administration bypasses first-pass metabolism. Each newly observed impurity above the identification threshold is isolated and tested for Ames activity using OECD 471 before batch release. The validated HPLC method uses UV detection below 210 nm, with column temperature held at 25 °C and a mobile phase containing a volatile buffer if LC-MS confirmation is needed. For oral solid dosage forms, the same method is used with a dissolution sample preparation step; the dissolution sample is injected after filtration through a 0.45 µm PVDF syringe filter, and the filter is checked for API adsorption below 2.0% because hydrophilic APIs can bind to nylon membranes.

Packaging component selection is also governed by ICH Q1A photostability and moisture protection data. PVC/PVDC-aluminium blisters are used for tablets, HDPE bottles with molecular sieve for capsules, laminated aluminium sachets for granules, and Type I glass with bromobutyl stoppers for injections. For each packaging configuration, the stability protocol includes assay, related substances, dissolution or reconstitution time, moisture, pH, and microbial limits. If the 6-month accelerated data at 40 °C/75% RH show more than 5% content loss or any unspecified degradation product above the qualification threshold, the package is re-evaluated before the product is assigned a shelf life. The terminal-products matrix therefore consists of immediate-release tablets, hard capsules, granules for oral suspension, oral solution, and injectable solution or lyophilised powder, each with its own excipient compatibility profile and ICH-derived control strategy.